New work computers (tech refresh), a conversation looking for ideas

[jpyeron]

We used to get high end (6-8k$ budget) laptops for each engineer. But recently laptops have been underperforming and in conflict with our requirements (e.g. no more actual sleep so they melt or die in transit).

Most of our engineers create VMs. Some use various graphics intensive software (Autodesk, kicad, Adobe Premiere, etc.)

This is a screenshot from the previous laptop refresh , it has 2 4TB SSD, but at reboot I am already using about 32G ram, and several times this month I ran out of ram.

So my new thought is, since our laptops have an always on VPN anyways, to split the workstation in 2.

Something like Latitude 9440 2-in-1 for 3k$, leaving another 3k$ for a desktop that is much more powerful than a laptop can ever be. But the last time I designed such a system Computer Shopper was 1.5" thick and in tabloid format.

Note some things require Windows, but I think the desktop could run Linux.

Ideas?

[jonpaul]

Lunix will choke on a lot of business apps and files.

Not for the novice users in an office.

We had HP Elitedesk an ProDesk WS, very reliable and flexible for WS use. $200..800

After HP, Dell went to China junk parts, we migrated to Asus laptop Zenbook (fails dead screens) and finally Lenovo Carbon X1, preferred by pro road warriors.

$1..2K

j

[jpyeron]

Lunix will choke on a lot of business apps and files.

Not for the novice users in an office.

These are not for novices.  But, that is why they will have a windows laptop too.

We had HP Elitedesk an ProDesk WS, very reliable and flexible for WS use. $200..800

Looking at the HP Elite Tower 600 G9 PC, maxed out at 128GB ram and 16 cores - not really impressive for 3k$

Was hoping to get closer to 32 cores and 256GB ram for the desktop paired with the lower end laptop (e.g. Latitude 9440 10 core, 32GB ram

[audiotubes]

For the price you mentioned, and if you need a laptop specifically, I can recommend Eurocomm. I have their Tornado F5W which at the time was the fastest laptop you could get. Some of the things to like about their laptops are that in most cases you get a socketed CPU so when this year's latest and greatest is no longer fast enough, you can make the change (as long as nextgen CPU has the same socket). They have good cooling, nVidia graphic cards etc, a fair bit of expansion (for example in my laptop, 3 disk drives). Mine has an i7777K but you can even get Xeons.

I work for a large company, we get high end Dells for work.  They cost a lot of money and are not very good values, especially compared to the Eurocomm units. We get 32G and it's enough for use but things are getting worse and more would be better for a lot of people. For me it's mostly about how many PDF manuals I can have open at one time. Usually 10-20 depending on the day.

Now that Windows has WSL (pretty much a real Linux running inside Windows) I feel the need to throw my Windows boxes off the side of a mountain much less. It's pretty ok. You can run quite a lot of Linux apps perfectly fine, some run even better than the Windows version. For example Visual Studio Code can be installed from inside WSL Linux and it is then available in Windows as well. I don't write code that runs on PCs but some of my coworkers do, and they work pretty much exclusively from inside WSL all day long.

I kindof agree with @jonpaul that Linux as a desktop is not yet a good solution in the grand scheme of things. Really depends on your company and requirements. If you guys are writing cross platform code then Windows with WSL is a very good option. If it's just apps, then it depends if your apps run on Windows, Linux, or both. And also how much of an IT department you have and what they will support.

As I'm typing I saw you mentioned towers. I have a G9 Business PC and didn't like it much. I bought a Lenovo Ryzen 7 Pro which I like a lot better. Anyway, look into small office servers, you often get a lot more for your money and they make perfectly good desktops. Most of them nowadays have onboard graphics in case that helps, and they have ECC RAM and Xeons etc. for less than a so-called tower desktop. I have Fujitsu and Dell office servers, I prefer the Fujitsu which is really well thought out with disk trays that are easy to deal with and other internal niceties.

[jpyeron]

I struggle at communications, so I will try to explain differently.

Our current old, out of warranty, laptops are 16 Core, 64GB ram, with 2x 4TB storage. They don't do a good job of meeting our needs.

I am thinking about issuing 2 devices to each engineer - a modest laptop and a super desktop. In aggregate it should still be within the typical 6-8k$ budget.

I am assuming that Laptops are going to continue to diverge on the usefulness vs cost factors. E.g. 1, cannot commute from home to work without loosing my work since Windows and modern hardware only do hybrid sleep now. 2, we are trying to do more than is reasonable on a laptop - I have many VMs.

I remember in 2006, we gave poweredge servers to each engineer at their desk. (fatal flaw - noise)

I think we can spend 2-3k$ on a laptop that we can standardize across the Company, and provide the engineers with a second machine for the heavy workloads. That laptop would be the "interface" to their "server" at their desk.

I guess I would be wrong on my "no laptop solution exists" assumption; if for 8k$ we could get a laptop with 128-256G of ram, that actually sleeps, has ~32 cores, with a next business day support plan.

For the price you mentioned, and if you need a laptop specifically, I can recommend Eurocomm. I have their Tornado F5W

Lower specs than what is currently issued to my engineers.

I work for a large company, we get high end Dells for work.  They cost a lot of money and are not very good values, especially compared to the Eurocomm units. We get 32G and it's enough

I own a small company - and we have been getting higher end Dells.

Now that Windows has WSL

A strongly worded "meh" - it has lots of issues playing nice, most of the time VMWare Workstation is a better solution, especially since WSL 2.

Lenovo Ryzen 7 Pro ... Fujitsu ... Dell office servers.

Worried about noise when buigin servers, and the "commercial" desktop solution seem to max out on the low end of specs. Again, this is why I am in a similar situation to https://www.reddit.com/r/sysadmin/comments/vd1igl/developer_workstation_that_needs_32_cores_and/

[nctnico]

Dell has precission workstations which are exactly for this purpose. Try to go for the lowest spec GPU you can because that will have the least amount of fan noise. But even with that, these systems are extremely quiet. Because I don't do anything with heavy 3D, I have fitted a fanless NVidia card in my Dell workstation.

[jpyeron]

Dell has precision workstations which are exactly for this purpose.

11,208.92$ for Precision 7865@32 core, 256 GB ram. Double the budget

Have IT build it like days of old?:

2500$ https://www.microcenter.com/product/674278/amd-ryzen-threadripper-7970x-storm-peak-40ghz-32-core-str5-boxed-processor-heatsink-not-included
0600$ https://www.microcenter.com/product/675036/gigabyte-trx50-aero-d-amd-str5-eatx-motherboard
0450$ https://www.amazon.com/Crucial-PC4-21300-2666MHz-CT32G4RFD4266-Registered/dp/B07X1TMZBS
0300$ 4TB SSD, more HD by project needs
0500$ Case, PS, misc (note good GPU video is on laptop...)
3000$ Laptop
=====
7350$

This is just a quick google - did not verify / test, look at it too in depth.

[richardmartino]

Sounds like you have an extremely specialized use case here. Building something custom (desktop) like Threadripper/Radeon might be a good fit for the high throughput 3D applications. I can imagine that the Lenovo workstation laptops might work for the secondary device.

[nctnico]

Dell has precision workstations which are exactly for this purpose.

11,208.92$ for Precision 7865@32 core, 256 GB ram. Double the budget

Have IT build it like days of old?:

No. You'll end up with a noisy PC with crappy thermal management.
What you do when buying a Dell is to get a relatively cheap base model and add the extra bits yourself. NVME SSDs, memory and video card are much cheaper to buy somewhere else than Dell.

[The Soulman]

I'm using a dell 5420 (i5 1135/ 32gb ram) with a Sonnet egpu housing as a "dockingstation" over thunderbolt, originally I planned to put a RTX4000 in it but that wasn't neccesary for my use case (simple 3d modelling).
It's now used for the additional IO and has a Blackmagic design 4k Extreme video capture card in it, the i5 still manages ok when ingesting and displaying raw 4k footage.

Here it is: https://www.sonnettech.com/product/egpu-breakaway-box/overview.html

Could such a egpu work in your use case?
Than you could get a laptop without gpu, and even a efficient cpu to improve real-world performance by much less thermal throttling? (that and running a proper full-size gpu)

Either way get a laptop that has upgrade-able memory and storage, this used to be the default but not anymore. 

[abeyer]

Since it sounds like the desktops would largely be accessed remotely anyway, rather than issuing a separate single desktop machine to each user, you might consider desktop virtualization running on a pool of servers (either your own, or cloud based, or some combination of both depending on your needs.)

This can give the flexibility to provision large cpu/gpu/memory-heavy instances when they're needed, smaller ones when they're not, and suspend them entirely when they're not being used.

[Halcyon]

Since it sounds like the desktops would largely be accessed remotely anyway, rather than issuing a separate single desktop machine to each user, you might consider desktop virtualization running on a pool of servers (either your own, or cloud based, or some combination of both depending on your needs.)

This can give the flexibility to provision large cpu/gpu/memory-heavy instances when they're needed, smaller ones when they're not, and suspend them entirely when they're not being used.

You beat me to the punch! Depending on what the OP's workflow is, virtualisation is a very viable options and has advantages over end-user equipment. For example, it's scalable, can be made redundant and it's centrally managed.

I've started to virtualise our heavy workloads at work, rather than replacing our big, heavy forensic workstations. Having that kind of technology sitting on everyone's desk and basically doing nothing most of the time is wasteful.

[jpyeron]

Since it sounds like the desktops would largely be accessed remotely anyway, rather than issuing a separate single desktop machine to each user, you might consider desktop virtualization running on a pool of servers (either your own, or cloud based, or some combination of both depending on your needs.)

This can give the flexibility to provision large cpu/gpu/memory-heavy instances when they're needed, smaller ones when they're not, and suspend them entirely when they're not being used.

You beat me to the punch! Depending on what the OP's workflow is, virtualisation is a very viable options and has advantages over end-user equipment. For example, it's scalable, can be made redundant and it's centrally managed.

I've started to virtualise our heavy workloads at work, rather than replacing our big, heavy forensic workstations. Having that kind of technology sitting on everyone's desk and basically doing nothing most of the time is wasteful.

Thinking hard on that one, sounds very reasonable. Worried about what I have not thought of.

[ajb]

Seconding or thirding the virtualization idea.  I am currently typing this message on a workstation VM that I've been piloting and it's been working great. 

We recently set up our first "real" virtualization infrastructure, with a small vsphere cluster.  We're doing this on-prem, partly because we are limited to not-amazing ISPs that serve our building, partly to take advantage of our 10G+ LAN, and generally because it's easier for a lot of our applications. 

A few things I've learned as a relative IT amateur along the way in case they help (no particular order):


- The 'right' way to do GPU intensive desktop work is to use nVidia's vGPU system, but this is fuck-off expensive.  IIRC, it would be something like $20k for a couple of supported GPUs and first year's software licenses (yes you have to pay both vmware and nVidia every year!) for just a handful of users.  It does give a lot of flexibility in terms of scaling users vs GPUs, and easily moving VMs around within a cluster, if you have enough users to warrant those things, but we don't.



- So instead of doing that, we bought a handful of much cheaper Quadros and are using the standard PCI passthrough capabilities provided by our ESXi hosts to expose them to the workstation VMs.  The main downside to this is your individual workstations are tied to a GPU in a specific host, but this is not a problem for our workflow.  There are barebones servers available with a dozen or so PCIe x16 slots (although IME a lot of CAD applications will work fine with a GPU in an x4 or x2 slot) which is probably where we'd move as our need for that sort of thing grows.  One thing to keep in mind with GPUs, is you will be relying on video encode on the VM for a lot of the higher-performance remote desktop solutions, so there is a higher baseline GPU demand, especially with multiple/larger displays.  You also may need a client with decent hardware decode, depending on throughput you need.



- Some applications simply don't perform well with some remote desktop systems.  For example, our Solidworks users are perfectly happy with standard Windows remote desktop, but Altium is absolutely awful with RDP.  Something about the way it round-trips user input and graphics updates I think -- it gets really laggy and squirrely.  I tried out Parsec for a while, but even on our 10G LAN it would have terrible drops in image quality that would take minutes to recover, if they recovered at all.  They use a fairly conventional video codec that simply doesn't work well for CAD type applications.  It also had a few other annoyances, like occasionally needing to log into your remote VM some other way because you needed to refresh your Parsec log in on that machine, not locking one remote connection when another is established, etc.  I am currently using HP Anyware (formerly Teradici), which is more expensive, but purpose-built for workstation use and it generally works great.  I'm running 2x 4k displays and it's indistinguishable from running a local machine.  I can also run 2x 4k displays at home via VPN and even with our crappy upload speeds at the office it's really good -- occasional frame rate drops when large areas of the displays are updating, but it stays crystal clear on every frame instead of turning into a mess of artifacts.   



- USB passthrough is another tricky point if you need to support things like debug interfaces or USB instruments.  Even before we got into virtualization, I was messing with this so I could basically bring my workstation with me to the bench with all of the test equipment.  (That's super convenient, by the way, and I highly recommend it.  Even without virtualization, being able to go from my office to the lab and use all of the equipment there without having to log into a different computer, start up whatever applications I was using, etc, was really nice.)  Remote desktop apps will do keyboard/mouse, audio, maybe USB drives, but generally not much else, IME. 

There are a few software options for remote USB out there, and I tried several of them and found them to work fine.  A lot of them require buying a 'server' license for a certain number of USB devices, which can add up if you have a lot of users needing to redirect a handful of devices each.  I settled on Flexihub which uses a per-connection 'credit' system.  That would be annoying long term, but is cheaper and more flexible, and for the most part has worked fairly well.  They have an Android app, so once or twice I've walked around with a J-Link plugged into my phone, and my iPad RDPed into my computer to do firmware updates on equipment.

More recently, I've been using a Digi AnywhereUSB 8, and it's been absolutely rock solid.  They explicitly support USB hubs, so you can expand the number of connected devices quite easily, and it's handled every device/use case I've tried with it -- including a J-Trace doing streaming cortex trace at 100MHz.



- If you do any network development, getting direct network connections to VMs for testing also requires consideration, since you can't just plug an extra USB-Ethernet adapter in when needed.  We assigned a set of VLANs for development/testing, and our VM hosts expose each of those to each dev/test VLAN.  For the other end of the equation, we have a handful of 5-port PoE switches configured with an 'all VLANs' uplink port, and then four ports assigned to those VLANs.  We have PoE (almost) everywhere, so it's very easy to just grab one of the preconfigured switches, plug it in anywhere, and select/enable the corresponding network connection in the VM.  We rely on this a lot for testing and customer support -- since our product is used with third-party control software, we have a set of otherwise-clean VMs prebuilt with different versions of that software we can easily fire up to troubleshoot if a customer has a problem.



- Storage needs consideration here as well.  Hopefully you already have decent network file storage, but you may want a dedicated NAS/storage network specifically for VMs, since they will be hitting the storage a lot harder.  You can of course use internal drives in the VM Hosts, but that makes it slower/harder to migrate VMs and implement backups properly.  This doesn't need to be anything insane for a small number of VMs, there are ready-made boxes from Synology for example that can do appropriate RAID configurations and bonded 10G network links that will work fine.

[nightfire]

We have a handful of developers at my workplace and we also are using this route. Basically the normal developer has a HP Probook as his desk machine and to be able to work from homeoffice, if needed, and the CPU-intensive stuff like compiling and debugging C++/c# code is done in a VM on a big server. As the developers rarely compile everything at the same time, this approach works for us.
But in todays servers you usually scale with more cores in the CPU, so your application should support such an approach.

For desktop virtualization, I also would prefer VMWare Workstation, especially if you need access to atttached devices that have to be mapped in the VM.

[thm_w]

Build yourself or go with a prebuilt, eg: https://www.cyberpowerpc.com/system/AMD-Threadripper-Pro (check for reviews, cant remember if this company was any good).

I would avoid Dell as their motherboards + PSUs are almost always proprietary.

[SiliconWizard]

+1 for a Threadripper Pro-based workstation at the moment.

As to virtualization, only the actual use case can determine if this is a workable approach, and if this is going to save any dollar (which is dubious depending on the number of workstations, required processing power, and actual availability of someone to administrate all this, which is not necessarily obvious.)

[m98]

A bit of a side question as I have a Dell Precision Workstation at work that does perform quite well, are the proprietary Dell Workstation motherboards really inferior to the stuff you get from the likes of Asus or Gigabyte? Guess I would trust Supermicro to be superior, but all the others with their RGB-ultra-xtreme gamer stuff?

[nctnico]

A bit of a side question as I have a Dell Precision Workstation at work that does perform quite well, are the proprietary Dell Workstation motherboards really inferior to the stuff you get from the likes of Asus or Gigabyte?

Quite the opposite. The Dell workstation motherboards are designed to run utterly reliable under high load as the computer crashing costs serious money in lost time. On top of that the higher end Dell systems (like the precission range) have ducts in them to direct airflow to where it is needed and the proprietary motherboard supports that. Over the years I have had quite a few Dell systems and none got retired due to a malfunction.

Some people don't like proprietary motherboards because they can't replace them with a standard motherboard. But that is not the reason you buy a Dell computer; you get a Dell for the reliability and low noise due to the good airflow design.

[thm_w]

A bit of a side question as I have a Dell Precision Workstation at work that does perform quite well, are the proprietary Dell Workstation motherboards really inferior to the stuff you get from the likes of Asus or Gigabyte? Guess I would trust Supermicro to be superior, but all the others with their RGB-ultra-xtreme gamer stuff?

Not really inferior in terms of stock performance no. But if your mobo or PSU dies, finding a drop in replacement is much harder than a generic ATX board. Or if you wanted to upgrade (CPU/GPU), choices may be very limited.
If you care about overclocking, Dell motherboards won't have it in the BIOS. They often leave out basic features like XMP as well (ram speed). So you could end up with RAM that runs slower than it is capable of.

Asus/gigabyte offer a variety of feature levels. If it has mass amounts of RGB its likely overpriced.
I'll take gigabyte over Dell any day, but, I can go into a BIOS and upgrade it and choose some settings. Most people that buy a Dell don't want to touch anything inside their PC.

[ajb]

As to virtualization, [...] availability of someone to administrate all this

While there is definitely additional complexity in the initial setup, and it requires some specific knowledge to set things up properly, day-to-day admin of virtual machines is largely the same as with physical machines and actually much *easier* in several ways.  User logins, mapping drives/printers, deploying software, all of that is exactly the same.  Dealing with the machines themselves -- deployment, backup/restore, that sort of thing -- is much easier and faster with VMs than with real hardware, and the real hardware you do deploy matters a lot less if it's mostly just used as a remote desktop terminal.  If a VM is acting up, it's very easy to take a snapshot before doing any invasive or risky troubleshooting, or just deploy a replacement from a template in a couple of clicks in the management interface.  Not having to go out to a user's desk to see what the problem is, scrounge up (or go buy) a replacement machine, image it -- or worse, manually configure it because you couldn't get a replacement compatible with your stock image -- saves a significant amount of time and headache.  VM Hosts do become a concentrated point of failure, but the marginal cost per-user to run them on proper server hardware with some redundancy isn't bad, and at least in vSphere there's decent support for fast, live migration of guest machines between hosts.

Of course it all depends a lot on the specific needs of the operation and what sort of IT resources you already have, but for smallish outfits, the admin load is probably about the same for equivalent physical and VM environments.  Managed/hosted solutions are also an option, though of course they have their own tradeoffs.  For us, in-house virtualization has been an unequivocal win, but YMMV.